HHS Public Access

Author manuscript
Harv Rev Psychiatry. Author manuscript; available in PMC 2020 May 02.
Author Manuscript

Published in final edited form as:

Harv Rev Psychiatry. 2019 ; 27(5): 303–316. doi:10.1097/HRP.0000000000000232.

Current Understanding of Religion, Spirituality, and Their

Neurobiological Correlates
James I. Rim, MD, JD*,
Department of Psychiatry, Vagelos College of Physicians and Surgeons, Columbia University

Psychiatry, New York State Psychiatric Institute, New York, NY

Jesse Caleb Ojeda, BA*,

Author Manuscript

Department of Psychiatry, Vagelos College of Physicians and Surgeons, Columbia University

Divisions of Translational Epidemiology, New York State Psychiatric Institute, New York, NY

Connie Svob, PhD,

Department of Psychiatry, Vagelos College of Physicians and Surgeons, Columbia University

Divisions of Translational Epidemiology, New York State Psychiatric Institute, New York, NY

Jürgen Kayser, PhD,

Department of Psychiatry, Vagelos College of Physicians and Surgeons, Columbia University

Divisions of Translational Epidemiology, New York State Psychiatric Institute, New York, NY

Cognitive Neuroscience, New York State Psychiatric Institute, New York, NY

Author Manuscript

Elisa Drews, MRes,

Mailman School of Public Health, Columbia University

Section for Translational Psychobiology in Child and Adolescent Psychiatry, Department of Child
and Adolescent Psychiatry, Centre for Psychosocial Medicine, Heidelberg University

Youkyung Kim, BS, Craig E. Tenke, PhD†,

Department of Psychiatry, Vagelos College of Physicians and Surgeons, Columbia University

Divisions of Translational Epidemiology, New York State Psychiatric Institute, New York, NY

Cognitive Neuroscience, New York State Psychiatric Institute, New York, NY

Jamie Skipper, MA,

Author Manuscript

Department of Psychiatry, Vagelos College of Physicians and Surgeons, Columbia University

Divisions of Translational Epidemiology, New York State Psychiatric Institute, New York, NY

Correspondence: Dr. Myrna M. Weissman, Columbia University Vagelos College of Physicians and Surgeons and New York State
Psychiatric Institute, 1051 Riverside Drive, Unit 24, New York, NY 10032. Myrna.Weissman@nyspi.columbia.edu.
†Deceased.
*Dr. Rim and Mr. Ojeda contributed equally and have agreed to share first authorship.
Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of
the article.
Supplemental digital contents are available for this article. Direct URL citations appear in the printed text and are provided in the
HTML and PDF versions of this article on the journal’s Web site (www.harvardreviewofpsychiatry.org).
 Rim et al. Page 2

Myrna M. Weissman, PhD

Author Manuscript

Department of Psychiatry, Vagelos College of Physicians and Surgeons, Columbia University

Divisions of Translational Epidemiology, New York State Psychiatric Institute, New York, NY

Mailman School of Public Health, Columbia University

Abstract
Religion and spirituality (R/S) have been prominent aspects of most human cultures through the
ages; however, scientific inquiry into this phenomenon has been limited. We conducted a
systematic literature review of research on the neurobiological correlates of R/S, which resulted in
25 reports studying primarily R/S with electroencephalography, structural neuroimaging (MRI),
and functional neuroimaging (fMRI, PET). These studies investigated a wide range of religions
(e.g., Christianity, Buddhism, Islam) and R/S states and behaviors (e.g., resting state, prayer,
Author Manuscript

judgments) and employed a wide range of methodologies, some of which (e.g., no control group,
varying measures of religiosity, small sample sizes) raise concerns about the validity of the results.
Despite these limitations, the findings of these studies collectively suggest that the experience of
R/S has specific neurobiological correlates and that these correlates are distinct from non-R/S
counterparts. The findings implicate several brain regions potentially associated with R/S
development and behavior, including the medial frontal cortex, orbitofrontal cortex, precuneus,
posterior cingulate cortex, default mode network, and caudate. This research may suggest future
clinical applications and interventions related to R/S and various disorders, including mood,
anxiety, psychotic, pain, and vertiginous disorders. Further studies with more rigorous study
designs are warranted to elucidate the neurobiological mechanisms of R/S and their potential
clinical applications.
Author Manuscript

Keywords
electroencephalography (EEG); electrophysiology; event-related potential (ERP); functional
magnetic resonance imaging (fMRI); magnetic resonance imaging (MRI); neurobiology; positron
emission tomography (PET); religion; spirituality

INTRODUCTION
Religion and spirituality (R/S) have been prominent aspects of most human cultures through
the ages, from ritual treatment of skulls during the paleolithic period 500,000 years ago to
the modern age of science, when 90% of the world population today is involved in some R/S
practice.1 Religion and mental health care were closely related until recent times, with the
first mental hospitals being located in monasteries, where were run by priests.1 Beginning
Author Manuscript

with the Enlightenment of the seventeenth and eighteenth centuries, secular philosophical
and scientific understandings began to predominate, culminating with Sigmund Freud’s
conception of religion as a shared delusion in the early twentieth century.2 This conception
set the tone for the psychiatric view of religion in the West for much of the twentieth
century, which relegated religion to the province of the clergy.2 Psychiatry’s approach
toward religion has begun to shift in the recent decades, however, with more research
concerning R/S and mental health, including research showing a significant impact of R/S

Harv Rev Psychiatry. Author manuscript; available in PMC 2020 May 02.
 Rim et al. Page 3

on mood, psychotic, anxiety, and substance disorders and, given the impact of R/S on
Author Manuscript

patients, increased awareness of the need for training on R/S factors in psychiatry residency
programs.1

The potential associations between R/S and mental health have been increasingly studied
over the past several decades. Although the studies are varied and nuanced in their findings,
they have supported several positive outcomes related to mental health and overall well-
being. For example, religious beliefs and commitments have been shown to help people cope
with stressful life events,3 and they have been associated with lower levels of death anxiety,4
higher levels of life satisfaction,5,6 better adjustment,7 and lower odds of developing
depression and anxiety.8 These studies varied in the specific aspect of R/S assessed (e.g.,
belief, commitment, “religiosity,” “intrinsic religiosity,” religious activities, maturity), as
well as in their outcome measures (e.g., coping with stress, life satisfaction, depressive
symptoms, anxiety symptoms). A consistent model of religion’s or spirituality’s impact on
Author Manuscript

mental health therefore remains difficult to ascertain. Nonetheless, several of these earlier
studies suggest that belief in the personal importance of religion might improve depressive
symptoms.6–8 More recent research has continued to suggest a role for R/S in mental health.
For example, R/S beliefs have been shown to be clinically significant sources of coping;9
frequent attendance at religious services has been associated with lower rates of suicide;10
and frequent religious participation has been linked with a lower risk of mental illness in
adolescents.11

Research in our laboratory has been consistent with these studies and has focused primarily
on the personal importance of R/S, regularly evidencing an inverse association with risk for
depression, as well as protective neurobiological correlates. These findings, which include
cross-sectional studies12 and also prospective studies across several decades and generations,
Author Manuscript

13 show the risk of depression to decrease by up to 80% when compared to non-R/S

controls. Our structural neuroimaging studies have found high R/S importance to be
associated with increased cortical thickness in the parietal and occipital regions, which may
confer resilience to the development of depression.14 A functional neuroimaging study
reported that greater R/S importance was associated with decreased default mode network
(DMN) connectivity, suggesting protective neural adaptation in the DMN as persons at high
risk for depression have increased DMN connectivity.15 Finally, two electroencephalography
(EEG) studies have shown associations between R/S importance and greater posterior alpha
during rest,16,17 which has been associated with better pharmacological treatment response
for depression.18

Although the studies cited above comprise a substantial body of work concerning R/S and its
Author Manuscript

clinical and neurobiological correlates, to our knowledge no comprehensive review has

examined all research concerning R/S and its association with neurobiological measures.
This article seeks to fill that gap by presenting a systematic review of the literature,
including research concerning the association of R/S with EEG, functional neuroimaging,
and structural neuroimaging. A significant body of research has looked at practices such as
meditation and mindfulness that were historically associated with R/S but that have
increasingly been used more broadly and stripped of their R/S meaning.19 These studies on
meditation have found various neurobiological correlates during meditation, including

Harv Rev Psychiatry. Author manuscript; available in PMC 2020 May 02.
 Rim et al. Page 4

increased low EEG frequency power of theta and alpha bands and, in functional
Author Manuscript

neuroimaging scans, increased activation of frontal and subcortical regions deemed

important for attention and emotion regulation.19 The research about meditation and
mindfulness that does not expressly concern R/S is excluded from this review; that research
has already been reviewed in other articles,19–21 and studies suggest R/S and meditation are
distinct, with differences in their neurobiological correlates.15 The purpose of this review is
to provide an overview of the existing research into R/S (including traditionally Western and
Eastern religions) and its neurobiological correlates, to assess the current state of research in
this area, and to foster analysis and development of models concerning R/S and depression
as well as other psychiatric disorders.

Religion has been defined as “a system of beliefs and practices observed by a community,
supported by rituals that acknowledge, worship, communicate with, or approach the Sacred,
the Divine, Ultimate Truth, Reality, or nirvana.”22 By contrast, spirituality has had more
Author Manuscript

varied definitions, ranging from the original meaning tied closely to religion, including
religious lives that reflect the faith’s teachings (such as Mother Theresa or Mahatma
Gandhi), to the more recent broader and nonreligious applications of the term to include life
purpose and meaning, connection with others, peacefulness, comfort, and joy.1 The latter
definition has been criticized as being meaningless and tautological, with the consequence
that it raises significant methodological problems for research.1 For the purposes of this
article, the original meaning of spirituality as tied to religion—namely, to people whose lives
reflect the teachings of their faith—is used both for methodological utility and to maintain
the distinctiveness of the term.

METHODS
Author Manuscript

Search Procedure
This systematic literature review was conducted using the MEDLINE (PubMed) and
PsycINFO (Ovid) electronic databases. This review adhered to the Preferred Reporting
Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines.23 A protocol
document outlining the methods, inclusion/ exclusion criteria, and hypothesis of the review
was created prior to the search. Articles published between 1 January 1990 and 15 October
2017 were eligible for inclusion. Search criteria for both databases were “Religion OR
Spirituality OR Meditation OR Mindfulness” AND “Neurosciences OR Neuroimaging OR
Electroencephalography OR Electrophysiology OR MRI” in Medical Subject Headings
(MeSH) terms for all fields. The initial search criteria included the terms “meditation” and
“mindfulness” in order to retrieve R/S articles that may not have expressly noted “religion”
or “spirituality.” Studies were limited to peer-reviewed journals, English language, and
Author Manuscript

human studies. Formulation of the research question was streamlined using the Problem,
Intervention, Comparison, Outcomes, Study design (PICOS) approach. The targeted studies
were those that compared “experienced” meditators or religious identifiers with novice/
secular controls. Studies that used spirituality as an intervention were excluded. Cross-
sectional and within-study designs were included, whereas case studies were excluded.

Relevant outcomes consisted of neuroimaging measures, including magnetic resonance

imaging (MRI), functional magnetic resonance imaging (fMRI), positron emission

Harv Rev Psychiatry. Author manuscript; available in PMC 2020 May 02.
 Rim et al. Page 5

tomography (PET), EEG, and event-related potentials (ERPs). These neurobiological

Author Manuscript

measures were selected for this review as they comprise the main, and most common,
measures currently used in neuroscience research. MRI is a structural imaging modality that
produces a static image of body tissues based on the magnetic resonance of atoms in the
body. fMRI is a functional imaging modality that measures brain activity by detecting blood
oxygenation and flow changes during neural activity. PET is a functional imaging modality
that measures tissue/organ functioning by introducing and then scanning for a radioactive
drug that collects in areas with higher levels of activity. EEG is an electrophysiological
modality that records electrical activity of the brain through electrodes placed on the scalp.
ERP is a stereotyped electrophysiological response, measured by EEG, to a specific
cognitive, sensory, or motor event.

Studies in this review were assessed for inclusion at five different stages: (1) titles and
abstracts, (2) full-text articles, (3) a data extraction phase that affirmed pertinence to our
Author Manuscript

research question, (4) a dual quality-assessment check for bias and accuracy/precision of
reported data, and finally (5) reassessment of remaining studies for R/S relevance. Sorting
was performed by different authors and collaborators over each stage (JCO, JS, ED, BD,
KI).1 First-stage screenings (JCO, JS) were scored for each inclusion criteria as “met,”
“unclear,” or “not met.” Articles in which all criteria were either “met” or “unclear” were
included for the next stage. Any articles that clearly did not meet at least one of the criteria
were excluded.

This process was repeated for the remaining full-text articles by three raters (JCO, JS, BD).
Only articles that clearly met each of the inclusion criteria could progress to the next stage.
Disagreements were discussed among the raters until a unanimous agreement was reached.
A data extraction sheet was utilized in the third stage to analyze the study details, findings,
Author Manuscript

and overlap between the included articles (JCO, JS). A frequency table was also developed
(ED) to see the overlap of topics and categories that the articles covered, with 13 categories
in total. Articles were categorized by the neuroimaging data recorded.

A quality assessment was utilized in the fourth stage to check for bias and to affirm the
integrity of the data collected from the studies. The remaining full-text articles were split up
between raters (JCO, JS, ED) and read once again. This assessment checked for (1) study
design, (2) funding bias, (3) data collection methods, (4) dropouts, and (5) appropriate
background literature. Studies that were determined to be of proper quality, as defined
above, were included in the final set. Additionally, our in-group specialists on EEG (JK,
CET) and MRI, respectively, reexamined the methods of data collection and experimental
design to determine the accuracy and precision of the data in each article. Studies that were
Author Manuscript

determined to be poor in quality were excluded.

After the initial search and review process, an additional exclusion criterion was added
because a large majority of articles related only to meditation and not to R/S, which is the
principal focus of this review.

1Baxter DiFabrizio and Katherine Iles, though not authors, assisted with this process.

Harv Rev Psychiatry. Author manuscript; available in PMC 2020 May 02.
 Rim et al. Page 6

Inclusion and Exclusion Criteria

Author Manuscript

The following criteria were used to determine the appropriateness of each article for the
study. All five had to be met to be included. These inclusion criteria were the following: (1)
all subjects must have been 18 years or older by the time of their assessment in each
respective study; (2) total sample size was n ≥ 10 (e.g., no case studies); (3) meditation/
spirituality/religion were used in a way relating to “experience” or “identity”; (4) original
data needed to be reported (e.g., no reviews, meta-analyses); and (5) articles were deemed to
be appropriate and accurate regarding quality assessment.

Additionally, two exclusion criteria were used to prioritize the focus of our review. The
overwhelming majority of the literature on this subject focuses on spirituality as an
intervention that teaches meditation skills to subjects or that introduces them to prayer/
religious practice. Furthermore, the initial search yielded predominantly articles concerning
Author Manuscript

meditation without a nexus to R/S, the latter being the focus of our review. To control for
these articles and focus on the morphological and activation differences in experienced
religious people, the following exclusion criteria (EC) were also used: (EC1) articles used
their spirituality component as an intervention with novice subjects who had little to no prior
experience with religion/meditation, and (EC2) articles studied meditation only and did not
expressly concern R/S.

RESULTS
Search Results
The search of databases resulted in a total of 1110 hits, consisting of 495 articles listed on
PubMed and 615 listed on PsycINFO (Supplemental Figure 1, available as supplemental
digital content at http://links.lww.com/HRP/A100). Of these, 148 reports were excluded on
Author Manuscript

the basis of duplication between databases. In the first stage of abstract screenings, 773
articles were excluded for not meeting one or more of the inclusion criteria or meeting EC1.
Following the full-report screening, 49 additional articles were excluded for not meeting
inclusion criteria or meeting EC1. Afterward, during the creation of the data extraction
sheet, 13 more articles were excluded for not meeting inclusion criteria or meeting EC1. The
dual quality-assessment check for bias and accuracy/precision of reported data was
performed on our EEG articles, which led to excluding 13 additional articles. The initially
process resulted in maintaining 114 articles. However, as 89 of these concerned meditation
without any express R/S nexus (EC2), the final count was 25 articles determined to be
appropriate for this review (see Supplemental Table 1 for dual quality assessment of the 25
included articles, available as supplemental digital content at http://links.lww.com/HRP/
A101).
Author Manuscript

Study Characteristics
The 25 articles selected for this review comprised a wide range of neurobiological
measurements, religions, R/S measurements, R/S states and behaviors, and structural and
functional finding. Neurobiological measurements (with number of articles) included EEG
(7) (Table 1), fMRI (13) and PET (1) (Table 2), and MRI (5) (Table 3) (see Supplemental
Table 2 for more details, available as supplemental digital content at http://

Harv Rev Psychiatry. Author manuscript; available in PMC 2020 May 02.
 Rim et al. Page 7

links.lww.com/HRP/A102). The total is greater than 25 as one of the studies utilized both
Author Manuscript

MRI and EEG. Religions investigated in the articles (with number of articles) included
Christianity/Catholicism (14), Buddhism (2), Islam (1), various (7), and unspecified (1)
(Tables 1–3). R/S measurements (with number of articles) included R/S importance (5),
religiosity (4), religious zeal (1), intrinsic religiosity (1), R/S orientation (1), R/S belief (6),
R/S conviction (1), importance of prayer and bible (1), frequency of prayer (3), church
attendance (1), church membership/attachment (5), devotional actions (1), self-identification
(4), and experience as a Sufi whirling dervish (1) (Tables 1–3). The various scales of R/S
that were used include the Centrality of Religiosity Scale, Minnesota Multiphasic
Personality Inventory, Royal Free Interview for Spiritual and Religious Beliefs, Quest, and
Duke University Religion Index (Tables 1–3). The total exceeds 25 as many studies used
multiple R/S measurements. Additionally, 5 of the articles did not have any R/S
measurement: all of the study participants were nuns in two studies;24,30 all of the study
participants were part of an orthodox Christian community in 2 studies;32,36 and all Tibetans
Author Manuscript

were assumed to be Buddhists in 1 study.38 The articles studied various R/S states and
behaviors, including (with number of articles) resting state (8), judgments (semantic, trait,
statement, moral) (7), prayer (3), mystical experience (2), anxiety-provoking task (1), social-
conformity task (1), neurofeedback training (1), pain perception (1), observing symbols (1),
and religious recitation (1) (Tables 1–3) Finally, the studies have found structural and
functional differences in a wide range of brain regions associated with R/S states or behavior
(Tables 1–3).

Overview of Study Findings

ELECTROENCEPHALOGRAPHY—Two studies associated EEG findings with personal
R/S importance and familial risk for depression. In Tenke and colleagues (2013),16 clinical
Author Manuscript

evaluations and self-reports of R/S importance were obtained among participants at high and
low risk of depression and again at ten-year follow-up when EEG was measured. They
found that greater posterior alpha was shown among participants who, during initial
assessment, reported R/S as highly important versus those who did not, that greater alpha
was shown among participants who reported R/S important in both sessions versus those
whose ratings changed, and that low alpha was shown among participants who expressed
R/S importance only at a later time. They concluded that posterior alpha may be a marker
for affective processes linked to depression, suggesting a possible association with the
development of spirituality. Tenke and colleagues (2017)17 extended the 2013 study to 20
years after the initial assessment. They found that greater posterior alpha was shown among
those who initially reported R/S as highly important versus those who did not, even if the
increase in their R/S rating occurred later, that low alpha was shown among those who
Author Manuscript

expressed R/S importance only at a later time, and that decreased alpha was shown among
those who changed their religious denomination. They suggested that there may be a critical
stage in the development of R/S and that it may be associated with posterior resting alpha.
This interpretation is supported by findings showing that resting posterior EEG alpha is
stable over long periods for adults, which is suggestive of an individual trait.45

Three studies recorded EEG while religious and nonreligious participants engaged in various
tasks. In Inzlicht and colleagues (2009),25 participants engaged in an anxiety-provoking

Harv Rev Psychiatry. Author manuscript; available in PMC 2020 May 02.
 Rim et al. Page 8

Stroop task while neural reactivity in the anterior cingulate cortex (ACC) was measured on
Author Manuscript

EEG. They found that error-related negativity localized to the ACC was decreased among
those with religious conviction, indicating reduced ACC activity. They concluded that
religious conviction affords a framework for acting within one’s environment and for
understanding one’s environment, which functions as an anxiety buffer and minimizes
experiencing error. In Thiruchselvam and colleagues (2017),27 participants engaged in a
social-conformity task in which they rated the attractiveness of faces after being provided a
peer rating. They found that the late positive potential, a centroparietal ERP component
considered to reflect emotional arousal (e.g., highly arousing pleasant or unpleasant
pictures) that is modulated by cognitive reappraisal, was altered by peer ratings in the
religious group but not in the nonreligious group. They concluded that, although both
religious and nonreligious groups on the self-report level yielded to conformity pressures,
the lack of such altered neural responses in nonreligious individuals suggested an association
between social conformity and religiosity. In Fondevila and colleagues (2012),26 participants
Author Manuscript

engaged in a semantic judgment task in which they judged religious and nonreligious
counterintuitive sentences. They found that the N400 ERP component, a centroparietal
negative wave linked to semantic expectancy,46 was smaller for religious than for
nonreligious counterintuitive ideas. They concluded that the mind processes religious ideas
as being more intuitive/plausible than nonreligious ones. Another explanation may be that
one’s expectations for the intuitiveness/plausibility of religious ideas might be lower.

The final two studies examined the association of religious practices, including prayer and
mystical experience, and quantitative EEG. In Kober and colleagues (2017),28 participants
underwent sensorimotor rhythm–based neurofeedback training. They found that individuals
who reported a high frequency of prayer showed improved neurofeedback performance
(increased sensorimotor rhythm/theta ratio) than those who reported a low frequency of
Author Manuscript

prayer. They concluded that due to their regular spiritual practice, the high-frequency prayer
group might have greater skillful in avoiding task-irrelevant thoughts and in gating
neurofeedback-system information. In Beauregard and colleagues (2008),24 while
quantitative EEG measures were obtained, Carmelite nuns underwent a mystical experience
by recalling and reliving their most intense mystical experience. Compared to the control
condition, in which the nuns recalled an intense experience with another person, they found
increased theta power over left and central frontal and parietal regions, greater theta/beta
ratio over frontal, central, temporal, and parietal regions, increased gamma1 power over
right temporal and parietal regions, increased theta connectivity (as assessed by EEG
coherence) between left frontal and central areas, and enhanced long-distant alpha
connectivity between right frontal, temporal, and parietal regions and between right central
and parietal regions. They concluded that mediation of mystical experiences occurs via
Author Manuscript

significant changes in EEG coherence and power.

Five of the seven EEG studies examined R/S behaviors (e.g., engaging in a mystical
experience or a judgment), whereas only two examined R/S states (e.g., those who report
R/S as being important vs. others) in which the participants did not engage in any activity.
While six of the seven EEG studies used a measure of R/S, the measures they used varied
widely (e.g., R/S importance, religiosity, frequency of prayer). One of the studies, namely
Beauregard and colleagues (2008),24 did not report using a measure of R/S while studying

Harv Rev Psychiatry. Author manuscript; available in PMC 2020 May 02.
 Rim et al. Page 9

the mystical experience of nuns; the religiosity of the nuns was assumed rather than
measured. One of the studies, Fondevila and colleagues (2012),26 measured R/S but did not
Author Manuscript

report associations of the measure with the EEG results, which could have provided
additional insight into the impact of religiosity on semantic judgment. Several of the EEG
studies employed mostly Christian/Catholic participants versus secular controls, though
some of the studies did not specify the exact R/Smakeup of the participants, which could
mask differences among various religious traditions.

FUNCTIONAL NEUROIMAGING (FMRI/PET)—Functional neuroimaging studies have

examined several religious practices, including mystical experience, prayer, and reciting
religious texts. In Beauregard and colleagues (2006),30 which preceded Beauregard and
colleagues (2008)24 and studied the same phenomenon, Carmelite nuns, while undergoing
fMRI scans, underwent a mystical experience by recalling and reliving their most intense
mystical experience. They found activation in “right medial orbitofrontal cortex, right
Author Manuscript

middle temporal cortex, right inferior and superior parietal lobules, right caudate, left medial
PFC [prefrontal cortex], left anterior cingulate cortex, left inferior parietal lobule, left insula,
left caudate, left brainstem, and extra-striate visual cortex.” They concluded that mystical
experiences are mediated by multiple brain systems and regions. Schjoedt and colleagues
examined various aspects of prayer, including engaging in high- and low-structured
prayer32,36 and listening to intercessory prayer.39 They found greater activation of the right
caudate nucleus in both high-and low-structured prayers versus comparable secular recitals,
which was seen as supporting the hypothesis that, as a type of recurring behavior, religious
prayer can stimulate the dopaminergic reward system in practicing individuals.32 They
found that improvised praying led to strong activation in the temporopolar region, medial
PFC, temporoparietal junction, and precuneus, from which they concluded that praying to
God is akin to normal interpersonal interactions.36 Finally, they found that Christian
Author Manuscript

participants deactivated the frontal network, including bilateral medial and dorsolateral PFC,
when listening to people who, they believed, possessed healing abilities, and that this
deactivation predicted their ratings of the speakers’ charisma and presence of God in prayer,
which suggested to the researchers a mechanism of authority that might facilitate
charismatic influence.39 In Azari and colleagues (2001),29 participants recited religious texts
while undergoing PET scans. They found that subjects who identify as being religious
activated a frontal-parietal circuit consisting of the dorsolateral prefrontal, dorsomedial
frontal, and medial parietal cortex. They noted that these areas perform a significant role in
maintaining reflexive evaluation of thought and concluded that religious experience may be
a process that is cognitive and that also feels immediate.

Many of the fMRI studies examined judgments of various kinds, including truthfulness of
Author Manuscript

statements, moral dilemmas, and attributing traits to self versus others. In Harris and
colleagues (2009),35 participants judged the truthfulness of religious and nonreligious
statements while undergoing fMRI scans. They found that judgment of religious statements
showed increased activation, compared to nonreligious statements, in the posterior cingulate,
precuneus, anterior cingulate, frontal pole, anterior insula, middle frontal gyrus, lateral
occipital gyrus, intraparietal gyrus, ventral striatum, inferior frontal gyrus, superior frontal
gyrus, thalamus, and cerebellum. They concluded that thinking about ordinary facts relies

Harv Rev Psychiatry. Author manuscript; available in PMC 2020 May 02.
 Rim et al. Page 10

more on memory retrieval net works, where as thinking about religious matters relies more
Author Manuscript

on brain regions that govern emotion, self-representation, and cognitive conflict.

In Christensen and colleagues (2014),40 participants judged moral dilemmas, including

deontological, utilitarian, personal, and impersonal moral dilemmas. They found that (1)
enhanced activity occurred in the precuneus, posterior cingulate cortex, middle temporal
pole, and medial superior frontal gyri for Catholics, whereas for atheists, enhanced activity
occurred in the superior parietal gyrus, (2) Catholics utilized differing brain regions for
utilitarian (dorsolateral PFC, temporal poles) and deontological moral judgments
(precuneus, temporoparietal junction), whereas atheists did not (superior parietal gyrus for
both judgment types), and (3) Catholics versus atheists showed increased activation in the
dorsolateral PFC and posterior cingulate cortex while engaged in utilitarian moral judgments
about impersonal moral dilemmas and increased activation in the anterior cingulate cortex
and superior temporal sulcus while engaged in deontological moral judgments about
Author Manuscript

personal moral dilemmas. They concluded that norms transmitted through religious practice
and indoctrination may influence moral judgments.

Four studies were conducted by Chinese researchers concerning judgment of traits—namely,

whether an adjective describes self or others—among different religious groups.31,34,37,38
Han and colleagues (2008)31 found that self-referential processing showed increased
activation for nonreligious participants in the ventromedial PFC (VMPFC) but for Christian
participants in the dorsomedial PFC (DMPFC). They concluded that Christian beliefs result
in different neural activity—namely, weakened activity for self-related stimuli but enhanced
activity for self-referential stimuli. Ge and colleagues (2009)34 found differing functional
connectivity among Christian subjects; namely, the connectivity between the medial PFC
and posterior parietal cortex/precuneus was different between trait judgments of the self and
Author Manuscript

the government leader but not between trait judgments of the self and Jesus. They concluded
that neurocognitive processes related to religious leaders are modulated by Christian belief
and practice, with the consequence that trait judgment of Jesus employs semantic trait
summary but results in decreased memory retrieval of behavioral episodes. Han and
colleagues (2010)37 found increased activation in the DMPFC/rostral anterior cingulate
cortex, midcingulate, and left frontal/insular cortex but not the VMPFC among Buddhists
engaged in self-judgment, and they found self-judgment was associated with decreased
functional connectivity between the posterior parietal cortex and DMPFC versus judgment
about a political figure. They concluded that weakened neural coding of stimulus self-
relatedness in the VMPFC results from Buddhist doctrine of no-self, whereas enhanced self-
referential stimuli evaluation occurs in the DMPFC. Finally, Wu and colleagues (2010)38
found increased activation in the left anterior cingulate cortex and VMPFC in self-
Author Manuscript

processing versus other-processing conditions among Han Chinese participants, in contrast

to increased activation in the middle temporal gyrus in self-processing versus other-
processing conditions among Tibetan participants. They concluded that the minimal
subjective sense of “I-ness” found among Tibetan Buddhists explains the difference between
Han Chinese and Tibetans.

Three studies examined other phenomena, including pain perception, viewing symbols, and
default mode network functioning at rest. In Wiech and colleagues (2009),33 participants

Harv Rev Psychiatry. Author manuscript; available in PMC 2020 May 02.
 Rim et al. Page 11

assessed the intensity of electrical stimulation that they had experienced when observing
Author Manuscript

religious versus nonreligious images while undergoing fMRI scans. They found that for the
religious group—all practicing Catholics—contemplating religious images allowed them to
detach from the experience of pain, and that activation in the right ventrolateral PFC
increased in the religious group when they viewed religious images during electrical
stimulation. The authors concluded that religious belief could provide a framework that
engages known brain pain-regulation processes. In Johnson and colleagues (2014),41
participants viewed religious and nonreligious negative and positive symbols while
undergoing fMRI scans. They found that (1) participants viewing religious negative symbols
(versus neutral symbols) showed deactivation in the medial occipital areas and the inferior
parietal, temporal, and anterior cingulate cortex, (2) participants’ Quest scale scores, which
is an index of religion and spirituality, were associated with greater activity in the primary
visual cortex only for negative symbols, and (3) when viewing religious symbols,
Author Manuscript

participants’ scores on the Beliefs About God Assessment Form, a measure of the
adaptability of a person’s religious beliefs, were associated with greater activity in the
amygdala and insula. The authors concluded that an early-stage visual mechanism may
underlie the interaction between spiritual quest/ adaptive religious beliefs and the processing
of visual religious symbols, and that the emotional nature of a person’s beliefs interacts with
the emotional perceptions of symbols. In Svob and colleagues (2016),15 participants at high
and low familial risk for depression underwent fMRI scans to assess DMN connectivity.
They found that R/S importance was associated with lower DMN connectivity in the left
lateral parietal lobe in high-risk individuals and that R/S importance was not associated with
familial risk in the DMN-central executive network regions. They concluded that R/S
importance may support a neural adaptation that is protective in the DMN of individuals at
high risk for depression—an adaptation that is different from those achieved through
meditation-based therapies.
Author Manuscript

All but one of the functional neuroimaging studies used fMRI. The one study using PET—
namely, Azari and colleagues (2001)29—is the oldest study included in this review, which is
consistent with the increasing use of fMRI as the modality of choice in neuroscience
research. Unlike the EEG studies that tended to investigate states, 13 of the 14 functional
neuroimaging studies examined R/S behaviors (e.g., engaging in a mystical experience or a
judgment), while only one examined R/S states (i.e., those who report R/S as being
important vs. others) in which the participants did not engage in any activity (which is
unsurprising, as functional neuroimaging is intended to study brain activity). Indeed, even in
the one study that did not require volitional activity—namely, Svob and colleagues (2016)15
—the “activity” of the DMN was measured. While a majority of the functional
neuroimaging studies used a measure of R/S, the measures that they used varied widely
Author Manuscript

(e.g., R/S importance, Quest scale, MMPI). Four of the studies—namely, Beauregard and
colleagues (2006),30 Schjoedt and colleagues (2008),32 Schjoedt and colleagues (2009),36
and Wu and colleagues (2010)38—did not report using a measure of R/S, as participants of a
particular community (nuns, Lutheran community, Tibetans) were assumed to be religious.
Again, most of the functional neuroimaging studies employed mostly Christian/Catholic
participants versus secular controls, though two studies examined Buddhist participants,
which may affect the generalizability of the results to other religious traditions.

Harv Rev Psychiatry. Author manuscript; available in PMC 2020 May 02.
 Rim et al. Page 12

STRUCTURAL NEUROIMAGING (MRI)—Five studies have utilized structural MRI to

Author Manuscript

assess persons with strong connection to R/S compared to nonreligious controls. In

Kapogiannis and colleagues (2009),42 researchers associated various components of
religiosity with neuroanatomical variability on MRI. They found (1) an association between
experience of an intimate relationship with God/engagement of religious behavior and
increased right middle temporal cortex volume, (2) an association between experience of
fear of God and decreased left precuneus and left orbitofrontal cortex (OFC) volume, and (3)
an association between traits concerning pragmatism and doubting God’s existence and
increased right precuneus volume. They concluded that key aspects of religiosity are
associated with differences in cortical volume. In Miller and colleagues (2014),14 second-
and third-generation offspring of depressed and depressed probands (first generation)
underwent assessment of R/S importance and church attendance, and cortical thickness
assessment via MRI. They found (1) an association, independent of familial risk, between
Author Manuscript

R/S importance (but not frequency of attendance) and thicker cortices in the left and right
parietal and occipital regions, the mesial frontal lobe of the right hemisphere, and the cuneus
and precuneus in the left hemisphere, and (2) significantly stronger effects of R/S
importance on cortical thickness among the high-risk versus the low-risk group, especially
along the left mesial wall. They concluded that a thicker cortex, which is associated with
high R/S importance, makes individuals at high familial risk more resilient to the
development of depression, potentially by countering the cortical thinning that is associated
with increased risk for depression. In Pelletier-Baldelli and colleagues (2014),43 researchers
associated nonclinical psychosis with the OFC volume on MRI. They found that intrinsic
religiosity is increased in the nonclinical psychosis group, that these individuals show
volume decreases in bilateral lateral and medial OFC, and that OFC volume was significant
negatively associated with depressive/negative symptoms. They concluded that nonclinical
psychosis–related brain abnormalities may also heighten religiosity. In Kober and colleagues
Author Manuscript

(2017),28 discussed above with regard to the EEG portion of their neurofeedback study,
participants also underwent MRI at rest. They found a negative association between gray
matter volume in the left medial OFC and neurofeedback performance in the highprayer-
frequency group. They theorized that successful suppression of neurofeedback-irrelevant
cognitive processes might explain this negative association. Finally, in Cakmak and
colleagues (2017),44 Sufi whirling dervishes, who can spin continuously for a long time,
underwent MRI scans to explore their resistance against vertigo. They found thinner cortical
areas for these subjects in the hubs of the DMN (precuneus and posterior cingulate gyrus),
and also in the motion perception and discrimination areas, including the right dorsolateral
PFC, right lingual gyrus, left visual area 5/middle temporal, and left fusiform gyrus. They
theorized that Sufis’ prolonged whirling without vertigo or dizziness may be related to the
cortical networks involved in motion/body perception.
Author Manuscript

In contrast to the functional neuroimaging studies, all of the structural neuroimaging studies
examined R/S states (e.g., those who rate as high R/S versus others) in which the
participants did not engage in any activity, which is consistent with MRI’s capacity to image
brain structures but not processes. Although all of the studies used a measure of R/S, the
measures they used, again, varied widely (e.g., R/S importance, religiosity, Duke University
Religion Index). One of the structural neuroimaging studies assessed Christian participants

Harv Rev Psychiatry. Author manuscript; available in PMC 2020 May 02.
 Rim et al. Page 13

versus secular controls; one enrolled Sufi Muslim participants; and the remaining enrolled
Author Manuscript

participants from various faith traditions.

DISCUSSION
The studies reviewed for this article investigated a wide range of R/S phenomena in
disparate ways. Despite the limited number of studies and the significant variations in the
methodology and the findings, two conclusions may be drawn: first, human experience of
R/S has been linked to neurobiological correlates, and second, the neurobiological correlates
for R/S phenomena appear to be different from the neurobiological correlates for
comparable non-R/S phenomena that involve similar emotions or cognition. In all but four of
the studies,24,30,32,36 participants with increased R/S states (e.g., R/S highly important) or
R/S behaviors (e.g., prayer) were compared with non-R/S controls (e.g., R/S not important,
reading a secular text) involving similar emotions or cognition, and in each of these studies,
Author Manuscript

the R/S state or behavior showed neural activation or structures that differed from the non-
R/S controls. While the R/S phenomena may be conceptualized as a psychological,
behavioral, and social phenomenon that is not inherently distinct from non-R/S phenomena,
it also suggests that it is associated with its own specific neurobiological effects.
Nonetheless, it should be emphasized that the specific brain regions preferentially involved
in R/S states and behaviors are not limited in their role to R/S, as all of the brain regions are
also involved in various other states and behaviors. Additionally, despite this fundamental
commonality among the studies, the significant variations make a coherent synthesis of the
studies challenging.

R/S State
One useful way to categorize and synthesize the studies is by the R/S state/behavior studied.
Author Manuscript

All five of the structural MRI studies, one of the fMRI studies, and two of the EEG studies
examined participants, religious and nonreligious, in their resting state. Both Kapogiannis
and colleagues (2009)42 and Miller and colleagues (2014)14 showed thicker cortices among
participants who reported having an intimate relationship with God or reported R/S was
important to them. Kapogiannis and colleagues (2009)42 and Miller and colleagues (2014)14
suggest a significant role of the precuneus among R/S participants, as Kapogiannis found
decreased volume among participants who experienced fear of God, while Miller showed
increased volume among those who had high R/S importance. These findings suggest that
fear of God and R/S importance are opposing ends of a spectrum of attitude toward God.
Both Kapogiannis and colleagues (2009)42 and Pelletier-Baldelli and colleagues (2014)43
showed decreased OFC volume among those who fear God42 or have depressive/negative
symptoms,43 which suggests decreased OFC volume may be associated with depressive and
Author Manuscript

psychotic symptoms. While Miller and colleagues (2014)14 did not find a significant volume
change in the OFC, they found a thicker mesial frontal lobe, which is adjacent to the OFC,
among the high R/S importance group, confirming a vital role for this region for depression
and possibly for psychotic disorders. Kober and colleagues (2017)28 found a negative
association among the religious participants between the OFC volume and neurofeedback
performance—which supports OFC changes among the religious. Cakmak and colleagues
(2017)44 found thinner cortices in the DMN regions (including precuneus) and in the

Harv Rev Psychiatry. Author manuscript; available in PMC 2020 May 02.
 Rim et al. Page 14

motion-perception regions, possibly adding support to the role of the precuneus among the
religious. Svob and colleagues (2016)15 found that greater R/S importance in the high-risk
Author Manuscript

group is associated with decreased DMN connectivity, which is consistent with Cakmak’s
finding of thinner cortices in the DMN and suggests decreased DMN activity among the
religious. The two studies by Tenke and colleagues (2013, 2017)16,17 both found increased
posterior alpha among the religious, which is consistent with the MRI and fMRI studies
indicating R/S differences involving the occipital cortex. Overall, these studies suggest
important roles of the precuneus, OFC, mesial frontal lobe, and DMN among those for
whom R/S is important.

R/S Behavior
A large majority of the articles studied participants engaged in an activity, which included
religious practice (prayer, mystical experience, religious recitation, viewing religious
Author Manuscript

images) and nonreligious tasks (judgments, anxiety-provoking task, social-conformity task,

neurofeedback training) while undergoing functional neuroimaging or EEG. With regard to
religious practices, in the two studies by Beauregard and colleagues (2006, 2008),24,30
increased activity was noted during a mystical experience in a wide range of brain regions,
including the OCF and medial PFC, which were two of the regions found to have anatomical
changes among the religious in MRI studies. Schjoedt and colleagues (2009)36 found
increased activation in several brain regions, including the precuneus and medial PFC,
during prayer, which also are regions reported to show anatomical changes with R/S
importance. Azari and colleagues (2001)29 found increased activation in the medial frontal
cortex for participants engaging in religious recitation, which is also consistent with the MRI
studies. Beauregard and colleagues (2006)30 and Schjoedt and colleagues (2008)32 both
found greater activation in the caudate, which suggests importance of this region in mystical
Author Manuscript

experiences of which prayer may be an example. Observing religious image or symbols did
not result in greater activation in these regions, however, though various other regions were
noted to have increased or decreased activation.33,41 Multiple studies concerning participants
making various judgments showed functional changes in the medial PFC, precuneus, and
posterior parietal cortex,31,34,35,37,40 mirroring the MRI studies in which these regions were
found to have anatomical changes among the religious. Those studies and other judgment
studies26,38 also found activation, however, in various other regions that are not consistent
across the studies. Finally, two EEG studies using an anxiety-provoking task25 and a social-
conformity task27 found ERP differences among the religious, suggesting different
processing of meaningful stimuli. Despite the common regions of activation noted, it is
perhaps more interesting that the various studies on religious practice and nonreligious tasks
had disparate regions of activation that are not replicated in the other studies. These
Author Manuscript

differences may, in part, be due to the widely varying study designs, including various
religious practices and other tasks that may not have a common neurobiological mechanism.

Examining R/S Processing in Broader Context

The fact that R/S is observed through such a widespread distribution in the brain further
strengthens the notion that R/S is a multifaceted construct that should not be separated from
its biological, environmental, and social contexts. For example, most of the brain regions
that have been implicated in our review of R/S are also key components of large-scale neural

Harv Rev Psychiatry. Author manuscript; available in PMC 2020 May 02.
 Rim et al. Page 15

circuits (e.g., DMN, frontoparietal executive network, fronto-temporal-parietal network) that

Author Manuscript

subserve higher-order brain functions, such as emotion processing, empathy, self-

knowledge, and self-referential reflective activity. Examples include the following: mood-
induction stimuli have been shown to lead to increased cerebral blood flow selectively in the
superior PFC and precentral area (areas that were also implicated for R/S in several studies
reviewed here);47 BOLD signals in the superior PFC, middle/inferior temporal gyrus, and
inferior parietal cortex (regions that have also been associated with R/S) have been
associated with the cognitive appraisal of oneself and others;48 a high level of activity in the
middle frontal gyrus and middle temporal gyrus has also been shown to be essential for the
subjective feeling of empathy,49 whereas activity in the superior temporal cortex has been
predictive of altruistic behaviors;50 and the concerted activation of the parietal/temporal,
posterior cingulate, and medial prefrontal cortices has been involved in producing mind-
wandering states.51 Taken together, it is reasonable to speculate that these brain regions
Author Manuscript

represent access to a neural reserve that likely results from the process of neuroplasticity. A
greater neural reserve could, in turn, support an enhanced cognitive reserve that enables R/S
people to cope better with negative emotions, more readily disengage themselves from
excessive self-referential thinking (e.g., rumination), and ultimately be more resilient in the
face of various psychopathologies.

Limitations of Current Research and Directions for Future Research

Initially, the review is notable for the relative paucity of research concerning neurobiological
correlates of R/S phenomena, as only 25 articles were found in our search spanning almost
three decades. The 25 articles comprise a wide range of hypotheses, methodologies,
findings, and conclusions. A significant limitation of this review is that a number of the
studies employed methodologies that raise significant concerns about the validity of the
reported findings. Four reports24,30,32,36 studied only members of a religious community and
Author Manuscript

failed to include any nonreligious control groups, which does not allow for determination of
brain regions preferentially activated among the religious. While some of the studies used an
established measure of religiosity such as R/S importance, Centrality of Religiosity Scale,
MMPI, Quest scale, and Duke University Religion Index, most relied on self-identification
or used measures that had not been independently validated; it is therefore unclear what
aspect of religion was being measured, if any. The use of the large variety of R/S
measurements also renders comparisons across studies difficult, underscoring the need for a
common, practical, and widely accepted R/S measurement scale. One study26 measured
religiosity, but it was not associated with the neurobiological measurement, whereas another
study38 did not assess the religiosity of its Tibetan participants and instead presumed that
they were Buddhists. Furthermore, all studies had small or moderate sample sizes, ranging
from n = 12 to n = 104, with only three studies having n > 52.14,15,17 It should be noted,
Author Manuscript

however, that obtaining neurobiological measures with large samples requires considerably
more resources compared to the collection of survey, self-report, behavioral, or
epidemiological data. Nonetheless, these methodological limitations raise doubts about the
validity and reproducibility of the results and also about the conclusions drawn therefrom.

In order to advance the research into R/S and neurobiology, future research should
adequately address the following issues. First, more research is needed concerning the most

Harv Rev Psychiatry. Author manuscript; available in PMC 2020 May 02.
 Rim et al. Page 16

salient features of R/S that can be reliably measured. Given the wide range of R/S measures
Author Manuscript

currently in use, research validating these measures and exploring their common and
dissimilar features would significantly improve the ability to compare and synthesize
studies. Second, neurobiological studies of R/S should employ more rigorous methods that
can withstand scientific scrutiny (i.e., allow replication attempts). Future studies should
include a validated measure of R/S, measurement of R/S among the experimental group (i.e.,
religious participants), a control group of nonreligious participants or a nonreligious activity,
and association of the R/S measurement with neurobiological measurement, preferably
employing parametric or continuous measures rather than categorical approaches.
Employing larger sample sizes would reduce the likelihood of false positives. Furthermore,
future studies should build upon the existing literature by referencing the relevant research
and discussing how the findings of the new research adds to the existing body of knowledge,
thereby enabling disparate findings to be better integrated. As one step in this direction, and
Author Manuscript

to facilitate that process, the present review provides an overview of the research to date.

Conclusion
The human phenomenon of religion and spirituality has had significant impact on societies
throughout history, and with the advent of the scientific era, R/S has increasingly been the
subject of scientific inquiry. The 25 articles included in this review explored various aspects
of R/S state and behavior employing a wide range of methodologies, some of which raise
concerns about the validity of the results. Despite these limitations, the studies suggest that
the experience of R/S has specific neurobiological correlates and that the neurobiological
correlates for R/S phenomena appear to be different from the neurobiological correlates for
comparable non-R/S phenomena. The studies implicate several brain regions potentially
associated with R/S development and behavior, including the medial frontal cortex,
Author Manuscript

orbitofrontal cortex, precuneus, posterior cingulate cortex, default mode network, and
caudate. Furthermore, the studies suggest possible clinical applications of R/S, including for
depression,14,15,43 anxiety disorder,25 psychotic disorder,43 pain disorder,33 and vertiginous
disorder.44 R/S has the potential to significantly affect the pathogenesis and treatment of
these clinical conditions. The limited amount of data and the quality of the available data,
however, necessitate more research before the efficacy of clinical interventions can be
reliably assessed. Still, the findings of Miller and colleagues (2014)14 suggest that it would
be beneficial to employ R/S support for patients at high risk for depression, as high
importance of R/S may confer resilience to the development of depression. The studies
reviewed here warrant further research employing more rigorous study designs to elucidate
both the neurobiological mechanisms of R/S in humans and the clinical application of such
mechanisms through clinical trials.
Author Manuscript

Supplementary Material
Refer to Web version on PubMed Central for supplementary material.

Acknowledgments
Supported, in part, by John Templet on Foundation grant nos. 54679 and 61330, and National Institute of Mental
Health grant no. 2-R01-MH36197 (all to Dr. Weissman).

Harv Rev Psychiatry. Author manuscript; available in PMC 2020 May 02.
 Rim et al. Page 17

REFERENCES
Author Manuscript

1. Koenig HG. Research on religion, spirituality, and mental health: a review. Can J Psychiatry
2009;54:283–91. [PubMed: 19497160]
2. Huguelet P, Koenig HG, eds. Religion and spirituality in psychiatry. New York: Cambridge
University Press, 2009.
3. Larson R Thirty years of research on the subjective well-being of older Americans. J Gerontol
1978;33:109–25. [PubMed: 618959]
4. Spilka B, Hood RW, Gorsuch R. The psychology of religion: an empirical approach. Englewood
Cliffs, NJ: Prentice-Hall, 1985.
5. Butt DS, Beiser M. Successful aging: a theme for international psychology. Psychol Aging
1987;2:87–94. [PubMed: 3268197]
6. Van Haitsma K Intrinsic religious orientation: implications in the study of religiosity and personal
adjustment in the aged. J Soc Psychol. 1986;126:685–7. [PubMed: 3807339]
7. Koenig HG, Kvale JN, Ferrel C. Religion and well-being in later life. Gerontologist 1988;28:18–28.
8. Atkinson BE, Malony HN. Religious maturity and psychological distress among older Christian
Author Manuscript

women. Intl J Psychol Religion 1994;4:165–79.

9. Baetz M, Toews J. Clinical implications of research on religion, spirituality, and mental health. Can
J Psychiatry 2009;54:292–301. [PubMed: 19497161]
10. VanderWeele TJ, Li S, Tsai AC, Ichiro Kawachi I. Association between religious service
attendance and lower suicide rates among US women. JAMA Psychiatry 2016;73:845–51.
[PubMed: 27367927]
11. Chen Y, VanderWeele TJ. Associations of religious upbringing with subsequent health and well-
being from adolescence to young adulthood: an outcome-wide analysis. Am J Epidemiol
2018;187:2355–64. [PubMed: 30215663]
12. Miller L, Warner V, Wickramaratne P, Weissman M. Religiosity and depression: ten-year follow-up
of depressed mothers and offspring. J Am Acad Child Adolesc Psychiatry 1997;36:1416–25.
[PubMed: 9334555]
13. Miller L, Wickramaratne P, Gameroff MJ, Sage M, Tenke CE, Weissman MM. Religiosity and
major depression in adults at high risk: a ten-year prospective study. Am J Psychiatry 2012;
Author Manuscript

169:89–94. [PubMed: 21865527]

14. Miller L, Bansal R, Wickramaratne P, et al. Neuroanatomical correlates of religiosity and
spirituality: a study in adults at high and low familial risk for depression. JAMA Psychiatry
2014;71: 128–35. [PubMed: 24369341]
15. Svob C, Wang Z, Weissman MM, Wickramaratne P, Posner J. Religious and spiritual importance
moderate relation between default mode network connectivity and familial risk for depression.
Neurosci Lett 2016;634:94–7. [PubMed: 27717831]
16. Tenke CE, Kayser J, Miller L, et al. Neuronal generators of posterior EEG alpha reflect individual
differences in prioritizing personal spirituality. Biol Psychol 2013;94:426–32. [PubMed:
23998996]
17. Tenke CE, Kayser J, Svob C, et al. Association of posterior EEG alpha with prioritization of
religion or spirituality: a replication and extension at 20-year follow-up. Biol Psychol 2017;124:
79–86. [PubMed: 28119066]
18. Tenke CE, Kayser J, Manna CG, et al. Current source density measures of electroencephalographic
alpha predict antidepressant treatment response. Biol Psychiatry 2011;70:388–94. [PubMed:
Author Manuscript

21507383]
19. Rubia K The neurobiology of meditation and its clinical effectiveness in psychiatric disorders. Biol
Psychol 2009;82:1–11. [PubMed: 19393712]
20. Cahn BR, Polich J. Meditation states and traits: EEG, ERP, and neuroimaging studies. Psychol Bull
2006;132:180–211. [PubMed: 16536641]
21. Lomas T, Ivtzan I, Fu CH. A systematic review of the neurophysiology of mindfulness on EEG
oscillations. Neurosci Biobehav Rev 2015;57:401–10. [PubMed: 26441373]

Harv Rev Psychiatry. Author manuscript; available in PMC 2020 May 02.
 Rim et al. Page 18

22. Koenig HG. Medicine, religion, and health: where science and spirituality meet. West
Conshohocken, PA: Templeton Foundation, 2008.
Author Manuscript

23. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews
and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. J
Clin Epidemiol 2009;62:e1–34. [PubMed: 19631507]
24. Beauregard M, Paquette V. EEG activity in Carmelite nuns during a mystical experience. Neurosci
Lett 2008;444:1–4. [PubMed: 18721862]
25. Inzlicht M, McGregor I, Hirsh JB, Nash K. Neural markers of religious conviction. Psychol Sci
2009;20:385–92. [PubMed: 19291205]
26. Fondevila S, Martín-Loeches M, Jiménez-Ortega L, et al. The sacred and the absurd—an
electrophysiological study of counterintuitive ideas (at sentence level). Soc Neurosci 2012;7:445–
57. [PubMed: 22146056]
27. Thiruchselvam R, Gopi Y, Kilekwang L, Harper J, Gross JJ. In God we trust? Neural measures
reveal lower social conformity among non-religious individuals. Soc Cogn Affect Neurosci 2017;
12:956–64. [PubMed: 28338784]
28. Kober SE, Witte M, Ninaus M, et al. Ability to gain control over one’s own brain activity and its
Author Manuscript

relation to spiritual practice: a multimodal imaging study. Front Hum Neurosci 2017;11:271.
[PubMed: 28596726]
29. Azari NP, Nickel J, Wunderlich G, et al. Neural correlates of religious experience. Eur J Neurosci
2001;13:1649–52. [PubMed: 11328359]
30. Beauregard M, Paquette V. Neural correlates of a mystical experience in Carmelite nuns. Neurosci
Lett 2006;405:186–90. [PubMed: 16872743]
31. Han S, Mao L, Gu X, Zhu Y, Ge J, Ma Y. Neural consequences of religious belief on self-
referential processing. Soc Neurosci 2008;3:1–15. [PubMed: 18633851]
32. Schjoedt U, Stødkilde-Jørgensen H, Geertz AW, Roepstorff A. Rewarding prayers. Neurosci Lett
2008;443:165–8. [PubMed: 18682275]
33. Wiech K, Farias M, Kahane G, Shackel N, Tiede W, Tracey I. An fMRI study measuring analgesia
enhanced by religion as a belief system. Pain 2008;139:467–76. [PubMed: 18774224]
34. Ge J, Gu X, Ji M, Han S. Neurocognitive processes of the religious leader in Christians. Hum
Brain Mapp 2009;30:4012–24. [PubMed: 19507157]
Author Manuscript

35. Harris S, Kaplan JT, Curiel A, Bookheimer SY, Iacoboni M, Cohen MS. The neural correlates of
religious and nonreligious belief. PLoS One 2009;4:e0007272. [PubMed: 19794914]
36. Schjoedt U, Stødkilde-Jørgensen H, Geertz AW, Roepstorff A. Highly religious participants recruit
areas of social cognition in personal prayer. Soc Cogn Affect Neurosci 2009;4:199–207. [PubMed:
19246473]
37. Han S, Gu X, Mao L, Ge J, Wang G, Ma Y. Neural substrates of self-referential processing in
Chinese Buddhists. Soc Cogn Affect Neurosci 2010;5:332–9. [PubMed: 19620181]
38. Wu Y, Wang C, He X, Mao L, Zhang L. Religious beliefs influence neural substrates of self-
reflection in Tibetans. Soc Cogn Affect Neurosci 2010;5:324–31. [PubMed: 20197287]
39. Schjoedt U, Stødkilde-Jørgensen H, Geertz AW, Lund TE, Roepstorff A. The power of charisma—
perceived charisma inhibits the frontal executive network of believers in intercessory prayer. Soc
Cogn Affect Neurosci 2011;6:119–27. [PubMed: 20228138]
40. Christensen JF, Flexas A, de Miguel P, Cela-Conde CJ, Munar E. Roman Catholic beliefs produce
characteristic neural responses to moral dilemmas. Soc Cogn Affect Neurosci 2014;9:240–9.
[PubMed: 23160812]
Author Manuscript

41. Johnson KD, Rao H, Wintering N, et al. Pilot study of the effect of religious symbols on brain
function: association with measures of religiosity. Spirituality Clin Practice 2014;1:82–98.
42. Kapogiannis D, Barbey AK, Su M, Krueger F, Grafman J. Neuroanatomical variability of
religiosity. PLoS One 2009;4:e7180. [PubMed: 19784372]
43. Pelletier-Baldelli A, Dean DJ, Lunsford-Avery JR, et al. Orbitofrontal cortex volume and intrinsic
religiosity in non-clinical psychosis. Psychiatry Res 2014;222:124–30. [PubMed: 24746701]

Harv Rev Psychiatry. Author manuscript; available in PMC 2020 May 02.
 Rim et al. Page 19

44. Cakmak YO, Ekinci G, Heinecke A, Çavdar S. A Possible role of prolonged whirling episodes on
structural plasticity of the cortical networks and altered vertigo perception: the cortex of sufi
Author Manuscript

whirling dervishes. Front Hum Neurosci 2017;11:3. [PubMed: 28167905]

45. Tenke CE, Kayser J, Alvarenga JE, et al. Temporal stability of posterior EEG alpha over twelve
years. Clin Neurophysiol 2018; 129:1410–7. [PubMed: 29729597]
46. Kutas M, Hillyard SA. Reading senseless sentences: brain potentials reflect semantic incongruity.
Science 1980;207:203–5. [PubMed: 7350657]
47. Baker SC, Frith CD, Dolan RJ. The interaction between mood and cognitive function studied with
PET. Psychol Med 1997;27:565–78. [PubMed: 9153677]
48. Ochsner KN, Beer JS, Robertson ER, et al. The neural correlates of direct and reflected self-
knowledge. NeuroImage 2005;28:797–814. [PubMed: 16290016]
49. Hooker CI, Verosky SC, Germine LT, Knight RT, D’Esposito M. Neural activity during social
signal perception correlates with self-reported empathy. Brain Res 2010;1308:100–13. [PubMed:
19836364]
50. Tankersley D, Stowe CJ, Huettel SA. Altruism is associated with an increased neural response to
agency. Nat Neurosci 2007;10: 150–1. [PubMed: 17237779]
Author Manuscript

51. Hasenkamp W, Wilson-Mendenhall CD, Duncan E, Barsalou LW. Mind wandering and attention
during focused meditation: a fine-grained temporal analysis of fluctuating cognitive states.
NeuroImage 2012;59:750–60. [PubMed: 21782031]
Author Manuscript
Author Manuscript

Harv Rev Psychiatry. Author manuscript; available in PMC 2020 May 02.
 Author Manuscript Author Manuscript Author Manuscript Author Manuscript

Table 1

Electroencephalogram

Study Neuro- R/S measurement Religion type Activity studied Principal findings with R/S participants/tasks
Rim et al.

biological
measure

Beauregard & Paquette (2008)24 EEG None (all nuns) Catholic Mystical experience Increased theta, theta/beta ratio, gamma1, theta
connectivity, long-distant alpha connectivity

Inzlicht et al. (2009)25 EEG Religious zeal, belief in God Various Anxiety-provoking task Reduced activity in anterior cingulate cortex

Fondevila et al. (2012)26 EEG Religiosity (measured, but not Catholic Semantic judgment Smaller amplitude of N400 event-related potentials
associated, with EEG)

Tenke et al. (2013)16 EEG R/S importance Unspecified None Greater posterior alpha

Tenke et al. (2017)17 EEG R/S importance Various None Greater posterior alpha

Thiruchselvam et al. (2017)27 EEG Religiosity Various Social-conformity task Increased late positive potential in parietal lobe

Kober et al. (2017)28 EEG, MRI Frequency of prayer, religiosity Christian, EEG: neurofeedback EEG: increased sensorimotor rhythm/theta ratio
(Centrality of Religiosity Scale) nonreligious training MRI: negative association between left medial
MRI: none orbitofrontal cortex volume and neurofeedback
performance

EEG, electroencephalogram; MRI, magnetic resonance imaging; R/S, religion and spirituality.

Harv Rev Psychiatry. Author manuscript; available in PMC 2020 May 02.
Page 20
 Author Manuscript Author Manuscript Author Manuscript Author Manuscript

Table 2

Functional Neuroimaging

Study Neuro- R/S measurement Religion type Activity studied Principal findings with R/S participants/tasks
Rim et al.

biological
measure

Azari et al. (2001)29 PET Self-identification, member of Christian Religious recitation Activation of dorsolateral prefrontal, dorsomedial frontal, and medial
Christian community, history parietal cortex

Beauregard & Paquette fMRI None (all nuns) Catholic Mystical Activation in right medial orbitofrontal cortex, right middle temporal
(2006)30 experience cortex, right inferior and superior parietal lobules, right caudate, left
medial prefrontal cortex, left anterior cingulate cortex, left inferior
parietal lobule, left insula, left caudate, left brain stem, and extra-striate
visual cortex

Han et al. (2008)31 fMRI Self-identification, attached to Christian Trait judgment Increased activity in dorsal medial prefrontal cortex
Christian community, MMPI
questions on religion (including
importance of prayer and Bible)

Schjoedt et al. (2008)32 fMRI None (all are part of orthodox Christian Prayer Greater activation of right caudate nucleus in both high- and low-
Lutheran community) structured prayers

Wiech et al. (2008)33 fMRI Membership, attendance, prayer, Catholic Pain perception Greater activation in right ventrolateral prefrontal cortex with religious
devotional actions images

Ge et al. (2009)34 fMRI Self-identification, attached to Christian Trait judgment Functional connectivity between medial prefrontal cortex and posterior
Christian community parietal cortex/precuneus differentiated between trait judgments of the
government leader and the self, but not between trait judgments of
Jesus and the self

Harris et al. (2009)35 fMRI Dedicated Christians (beliefs, Christian Statement Increased activation in posterior cingulate, precuneus, anterior
convictions) judgment cingulate, frontal pole, anterior insula, middle frontal gyrus, lateral
occipital gyrus, intraparietal gyrus, ventral striatum, inferior frontal
gyrus, superior frontal gyrus, thalamus, and cerebellum for religious
statements

Schjoedt et al. (2009)36 fMRI None (all are part of orthodox Christian Prayer Increased activation in temporopolar region, medial prefrontal cortex,
Lutheran community) temporoparietal junction, and precuneus

Harv Rev Psychiatry. Author manuscript; available in PMC 2020 May 02.
Han et al. (2010)37 fMRI Self-identification, attached to faith Buddhist Trait judgment Increased activations in dorsal medial prefrontal cortex/rostral anterior
community, R/S importance cingulate cortex, midcingulate and the left frontal/insular cortex during
self-judgment
Decreased functional connectivity between dorsal medial prefrontal
cortex and posterior parietal cortex during self-judgment

Wu et al. (2010)38 fMRI None (Tibetans assumed to be Buddhist Trait judgment Stronger activation in middle temporal gyrus in self-processing
Buddhist)

Schjoedt et al. (2011)39 fMRI Prayer, belief in healing, persons with Charismatic Listening to Deactivation of medial and dorsolateral prefrontal cortex bilaterally
healing powers Christians intercessory prayer

Christensen et al. fMRI R/S beliefs (strong view, importance, Catholic Moral judgment Enhanced activity was in precuneus, posterior cingulate cortex, middle
(2014)40 external entity) using Royal Free temporal pole, and medial superior frontal gyri
Page 21
 Author Manuscript Author Manuscript Author Manuscript Author Manuscript

Study Neuro- R/S measurement Religion type Activity studied Principal findings with R/S participants/tasks
biological
measure
Interview for Spiritual and Religious
Beliefs
Rim et al.

Johnson et al. (2014)41 fMRI R/S orientation and belief using Various Viewing religious/ Deactivation in medial occipital, inferior parietal, temporal, and anterior
Quest scale nonreligious cingulate cortex while viewing religious negative symbols
negative and
positive symbols

Svob et al. (2016)15 fMRI R/S importance Various (mostly None Lower default mode network connectivity in left lateral parietal lobe in
Catholic) high-risk individuals

fMRI, functional magnetic resonance imaging; MMPI, Minnesota Multiphasic Personality Inventory; PET, positron emission tomography; R/S, religion and spirituality.

Harv Rev Psychiatry. Author manuscript; available in PMC 2020 May 02.
Page 22
 Author Manuscript Author Manuscript Author Manuscript Author Manuscript

Table 3

Structural Neuroimaging

Study Neuro- R/S measurement Religion type Activity studied Principal findings with R/S participants/tasks
Rim et al.

biological
measure
Kapogiannis et al. MRI Religiosity (multifactorial) Various None Intimate relationship with God and engaging in religious behavior associated
(2009)42 with increased volume of right middle temporal cortex
Experiencing fear of God associated with decreased volume of left precuneus and
left orbitofrontal cortex
Pragmatism and doubting God’s existence associated with increased volume of
the right precuneus

Miller et al. (2014)14 MRI R/S importance, Church Various None Importance of religion or spirituality (however, not attendance frequency)
attendance associated with thicker cortices in the left and right parietal and occipital regions,
mesial frontal lobe of the right hemisphere, and cuneus and precuneus in the left
hemisphere

Pelletier-Baldelli et MRI Intrinsic religiosity using Various None Bilateral volume reduction in both the lateral and medial orbitofrontal cortex
al. (2014)43 Duke University Religion Negative relationships between orbitofrontal cortex volume and depressive and
Index negative symptoms

Cakmak et al. MRI Traditional experienced Sufi Muslims None: history of Thinner cortical areas in precuneus and posterior cingulate gyrus, right
(2017)44 Sufi whirling dervishes whirling dervishes dorsolateral prefrontal cortex, right lingual gyrus, left visual area 5/middle
temporal, and left fusiform gyrus

Kober et al. (2017)28 EEG, MRI Frequency of prayer, Christian, EEG: neurofeedback EEG: increased sensorimotor rhythm/theta ratio
religiosity (Centrality of nonreligious training MRI: negative association between left medial orbitofrontal cortex volume and
Religiosity Scale) MRI: none neurofeedback performance

EEG, electroencephalogram; MRI, magnetic resonance imaging; R/S, religion and spirituality.

Harv Rev Psychiatry. Author manuscript; available in PMC 2020 May 02.
Page 23